Sponsored by
You are probably burning thousands of dollars on LinkedIn Ads right now, and you don’t even realize it
Your dashboard says your Cost Per Lead is dropping, and your marketing team is out celebrating.
But when your sales team actually calls those leads? It’s just college students doing homework and random vendors checking out the competition.
I call this the 'Fake Lead Tax,' and it is destroying B2B marketing budgets.
But what if you had an AI operator that caught this daily, separating the real buyers from the fake signal, before you wasted another dime?
Today, I'm giving you the exact, open-source AI agent that does exactly that.
Get it free (scroll to the bottom).
Sponsored
SaaS Billing Is Breaking. Here's How to Fix It.
Usage-based pricing is the new default — but most billing platforms weren't built for it. Get the Tabs Buyer's Guide to learn what modern revenue infrastructure actually requires.
The fake lead problem (and why LinkedIn makes it worse)
Most LinkedIn Ads teams don't have a cost-per-lead problem.
They have a buyer-quality problem.
Here's how it works. You launch a campaign targeting "Heads of Growth at SaaS companies with 50+ employees." Fine. Then LinkedIn's auto-expansion drifts the audience outward because it wants to spend your budget. Forms get optimized for "easy fill." CPL starts looking great. Leads start flowing.
Except half of them are:
students doing "market research"
vendors "checking out the competition"
solopreneurs at 3-person shops who will never qualify
enterprise folks who grabbed the deck and will never book a call
people who already have your competitor and just want the comparison PDF
Sales opens the lead. Sales closes the tab. Nobody follows up. The number on the dashboard is meaningless.
The math is brutal. LinkedIn clicks are $10-25 each. A "cheap" $150 lead is 10+ clicks of wasted spend on the wrong person. If 30% of your leads are fake, you're burning a third of your LinkedIn budget on people who will never buy.
Your dashboard celebrates. Your sales team rolls their eyes.
Same pattern I called out in the Outcome Kit, just LinkedIn-specific. And LinkedIn makes it worse because the clicks are 5 to 10x more expensive than Meta.
The buyer-quality layer never existed.
Nobody checked which leads turned into opportunities. Nobody mapped which message brought the CFO vs. the intern. Nobody decided which Thought Leader post was worth sponsoring based on whether buyers reacted, not whether coworkers liked it.
So the account optimized for the wrong thing. The report looked great. The pipeline stayed empty. And eventually someone on the exec team asked "why are we still spending on LinkedIn" and marketing didn't have an answer.
What I built
LinkedIn Ads Kit is an AI LinkedIn Ads manager that gives you a daily brief: what's real, what's fake, what's leaking, what to do next, and what not to touch yet.
A daily LinkedIn Ads operator with memory and a seatbelt.
The one-liner I keep coming back to:
Meta Ads Kit helps you find winners, bleeders, fatigue, and budget shifts.
LinkedIn Ads Kit helps you find real buyers, fake signal, sales leaks, and the next safest move.
It runs two ways:
Mode | What it does |
|---|---|
Export Mode | Zero API setup. Feed it LinkedIn Ads CSV exports, lead form exports, and your CRM. Daily brief or deeper audit in an afternoon. |
Connected Mode | OAuth into LinkedIn Marketing APIs. Pulls account data directly. Drafts safe changes. Applies only with explicit confirmation. |
Works with Claude Code, OpenClaw, Hermes, or any agent framework that reads skills.
See the Meta Ads kit in action:
The Daily Brief
Start here.
"Write me today's LinkedIn Ads brief."
"Run the LinkedIn daily check for Acme."
"What did my LinkedIn account do yesterday? Real, fake, leaks, what to do today."
The agent invokes the linkedin-ads skill, reads your brand memory, ingests CSV exports or pulls through the API, and hands back the brief.
If you live in the terminal, the CLI equivalent is:
npm run daily:brief -- --campaigns campaigns.csv --leads leads.csv --crm crm.csvOr the shorter version:
npm run daily-checkEither way, the brief is read-only by default. You don't have to let the kit touch your account. The brief is useful by itself.
Every brief answers five questions:
Are we buying real buyers?
What is creating fake confidence?
Where is the leak between ad, form, and sales?
What should we do today?
What should we avoid touching yet?
And it lands in six sections:
Section | What it means |
|---|---|
Real Signal | Spend that's creating qualified buyer conversations. |
Fake Signal | Spend that looks fine in-platform but lacks buyer-quality evidence. |
Leaks | Breaks between audience, offer, lead form, CRM, and sales handoff. |
Today's Moves | Specific next actions: pause, rewrite, inspect, score, or leave alone. |
Do Not Touch Yet | Campaigns where the data's too thin for confident action. |
Data Gaps | The exports, API permissions, or CRM notes the kit needs for a better read. |
The Four LinkedIn Ads States
The daily brief runs on a diagnosis model. Every campaign, every ad, every lead gets sorted into one of four states.
1. Real Signal
The campaign is producing evidence the right buyers care. Qualified leads. Opportunity-stage CRM matches. Comments from actual ICP buyers. Form answers that show real pain.
Operator move: protect the learning. Scale slowly. Create sibling angles. Document why it works.
2. Fake Signal
Campaign looks great in Campaign Manager. Produces weak downstream quality. Cheap CPL. Strong CTR. Students, vendors, tiny companies, job seekers. Sales rejection notes pile up.
Operator move: stop celebrating the dashboard. Tighten audience or form friction. Rewrite the offer around a specific buyer problem. Pause if quality doesn't recover.
3. Trust Gap
Audience and offer are probably right, but the ad has no credibility. Low CTR on company-page ads. Generic product claims. No proof. No human voice. Organic posts outperforming paid creative.
Operator move: test Thought Leader Ads. Add proof. Turn internal POV into external creative. Use a human voice when the company voice feels flat.
4. Handoff Leak
Ad created interest. Lead died after conversion. Sales accepted rate below expectation. Delayed follow-up. Reps lacking ad/form context. CRM stages that don't map back to the campaign.
Operator move: add hidden campaign context. Write sales handoff notes. Add one fit question. Map form answers to sales talking points.
That's the diagnosis. The daily brief is the output: Real Signal gets protected, Fake Signal gets challenged, Trust Gap and Handoff Leak both surface under "Leaks," and every state rolls into "Today's Moves" with a specific operator action.
LinkedIn is not Meta with expensive clicks
LinkedIn is not a volume channel. It's a trust and buyer-quality channel.
Meta rewards creative velocity, emotional hooks, and scroll-stopping patterns. LinkedIn rewards the opposite: the right buyer, hearing a credible point of view, taking a step that filters for fit.
That's why:
Company-page ads underperform. There's no face. No trust transfer. Logos don't carry conviction.
Thought Leader Ads work. Trust transfers from the person, not the company.
Broad audiences drift into garbage. Your buying committee is narrow. LinkedIn optimizes for spend, not fit.
Cheap leads are a trap. The form that's easy to fill is the form students fill.
LinkedIn Ads Kit is built around this belief. Every stage asks the same question: is this creating a real buyer conversation? Not "did we get a click." Not "is the CPL down." Is there a buyer on the other side who will actually convert?
The 10 skills
Skill | What it does |
|---|---|
| Core operator workflow. Loads brand memory, inspects data, produces the brief |
| OAuth, account discovery, API health, connected pulls |
| Safe draft review, dry run, apply, audit trail |
| Separates real buyer signal from fake lead volume |
| Maps lead and CRM records into fit, intent, and sales-readiness |
| Finds whether the offer is strong enough for LinkedIn traffic |
| Scores posts and builds API or Campaign Manager launch paths |
| Reviews lead forms for conversion friction AND qualification quality |
| Turns findings into an executive-ready operator brief |
| Creates sales context from campaign, offer, form, and lead data |
Every skill is a plain markdown file. Any agent framework that reads skills can run this. Each skill runs standalone or as part of the full pipeline.
Thought Leader Ads: the sharp edge
Thought Leader Ads are the one format on LinkedIn that consistently beats company-page creative. But most teams sponsor posts because the post got likes.
That's not enough.
This kit scores candidate posts on a buyer-quality rubric:
Signal | Question |
|---|---|
ICP fit | Is the right buying committee likely to care? |
Buyer pain | Does it name an expensive problem? |
Trust | Does it sound like lived experience, not brand copy? |
Organic signal | Did the market react before paid spend? |
Offer fit | Does it connect to a real next step? |
Sales usefulness | Would sales be glad this conversation started? |
If API creation is available, the kit drafts the creative action. If LinkedIn's approval mechanics block API creation, it produces a Campaign Manager launch packet: post URL, objective, approval copy, budget test, UTMs, measurement notes, what to watch before scaling.
Either way, you're not sponsoring the post because it got 12 likes from your team's company accounts. You're sponsoring it because the market already told you buyers care.
Safe Apply: the seatbelt
LinkedIn is expensive. A bad change at 2am costs real money. So the kit does not freehand the account.
The loop is:
READ → BRIEF → DECIDE → DRAFT → APPROVE → APPLY → LEARNYou stay in the driver's seat at DECIDE and APPROVE. The agent takes the grunt work on either side.
So what this looks like for you is something like:
"Draft a pause for the Generic Demo campaign. The reason is spend without buyer-quality signal. Include rollback notes."
"Dry run the pause draft against the live account."
"Okay, apply it."
Every apply starts as a markdown draft the agent writes for you. You read it. You approve it. Every draft validates against current account state before a write lands.
V1 supports: pause campaign, pause creative, set campaign daily budget, activate approved draft creative, create Thought Leader creative draft (or Campaign Manager fallback packet).
Still excluded: deleting entities, broad campaign creation, bid strategy changes, audience expansion, blind rebuilds, anything without brand/account validation.
You stay in the loop. The agent takes the drudgery. The account stays un-destroyed at 2am.
Multi-brand memory
If you run more than one brand, this is the part that compounds.
Each brand gets its own memory folder:
workspace/brands/acme/
├── profile.md
├── audience.md
├── voice-profile.md
├── offer.md
├── learnings.md
└── linkedin/
├── account.md
├── briefs/
├── drafts/
├── audit-trail/
└── cache/
The important file is learnings.md. Every brief and every apply run appends what the kit learns about the brand: who converts, what sales rejects, which offers create quality, which Thought Leader voices deserve spend, what not to repeat.
The agent gets smarter every run. The agency gets a durable knowledge base per client. The next LinkedIn Ads manager who rolls onto the account doesn't start from zero.
How this plugs into the rest of the stack
This is the eleventh kit I've open-sourced in 11 weeks (yeah, still counting, and still free). And like the rest, it's designed to fit into the full paid pipeline:
Sign up for StealAds early access here: https://stealads.ai
LinkedIn Ads Kit slots in as the third paid channel operator. If you're already running the Meta Ads Kit and the Google Ads Copilot, this is the missing one.
And if you're already running the Outcome Kit, LinkedIn gets its own channel-specific read instead of being lumped into a generic attribution view. Because LinkedIn is expensive and different. Buyer-quality logic that makes sense for Meta would make you look foolish on LinkedIn.
Who this is for
B2B founders running LinkedIn paid and tired of explaining cheap-CPL-big-waste to the board
Agencies managing multiple LinkedIn accounts who need a durable operator framework
Heads of growth at $1-20M ARR who know LinkedIn should work but haven't cracked the buyer-quality layer
Operators running Thought Leader Ads and looking for a repeatable way to decide what to sponsor
Not for you if you have a senior LinkedIn operator on payroll who is already mapping leads to CRM stages, scoring posts on ICP fit, and drafting account changes with audit trails. You don't need this. You're the 1%.
Get the kit
Full system free on GitHub:
You get:
The Daily Brief — read-only by default. Real Signal, Fake Signal, Leaks, Today's Moves, Do Not Touch Yet, Data Gaps. Ask your agent "write me today's LinkedIn brief" (or run
npm run daily:brief/npm run daily-checkfrom the CLI).10 skills covering the full pipeline from account read to safe apply
The Four LinkedIn Ads States framework: real signal / fake signal / trust gap / handoff leak, with a specific operator move for each
Two modes: Export Mode for zero-setup daily briefs, Connected Mode for API-backed drafts and apply
Flexible lead form CSV mapping — LinkedIn lets teams rename fields per form, so the kit maps common aliases and preserves unmapped columns. No reformatting required.
Safe Apply with dry runs, explicit confirmation, current-value validation, and full audit trail. READ → BRIEF → DECIDE → DRAFT → APPROVE → APPLY → LEARN.
Thought Leader Ads selector with an ICP / pain / trust / organic-signal / offer / sales-usefulness rubric, plus Campaign Manager fallback packets when API creation is gated
Multi-brand memory that compounds learnings across client accounts (the
learnings.mdfile per brand)Form friction review that looks at qualification quality, not just fill rate
Sales handoff generator that gives reps context for the conversation the ad actually started
Full CLAUDE.md, skill docs, install script, demo data, the works. Runs on Claude Code, OpenClaw, Hermes, or any agent framework that reads skills. MIT license.
No upsell. Fork it.
Stop paying LinkedIn a fake-lead tax. Start getting the daily brief Campaign Manager should have written for you.
Go big,
Matt
P.S. If you missed the last few kits, they're all still free and they plug into each other: Creator Breakout Kit, Outcome Kit, Landing Page Factory, ScrollClaw, Google Ads Copilot, Meta Ads Kit. LinkedIn Ads Kit slots in next to Meta Ads Kit and Google Ads Copilot as your third paid channel operator.
We've been running these agents for founders. Reply to this email and say 'AGENT' if you want on the list.
If this newsletter lit a fire under you, forward it to one person who would benefit.
🎁 Get rewarded for sharing! My team grew NEW accounts to over 50 million views in just a few months. We made an AI viral hook generator so you can follow the same hook frameworks that we do.
Invite your friends to join, and when they sign up, you’ll unlock our AI Viral Hook Generator—the ultimate tool for creating scroll-stopping content.
{{rp_personalized_text}}
Copy and paste this link: {{rp_refer_url}}